Development and validation of a 10-gene signature for predicting recurrence risk in HR+/HER2- early breast cancer undergoing chemo-endocrine therapy.

BACKGROUND

While existing multi-gene assays aid adjuvant treatment decisions, no gene signature has identified HR+/HER2- early breast cancer (EBC) patients at high recurrence risk post-chemo-endocrine therapy (C-ET).

METHODS

Clinical data and RNA sequencing information from 1457 HR+/HER2- breast cancer patients were collected from West China Hospital, the GEO database, and the TCGA database. Using univariate Cox regression, gene set enrichment analysis, and LASSO regression, ten key genes associated with recurrence were identified. A comprehensive prognostic model was developed by combining the 10-gene risk score with clinicopathological features, and a nomogram was created to predict 3-, 5-, and 7-year recurrence-free survival (RFS). The model's performance was evaluated using AUC and decision curve analysis (DCA).

RESULTS

The 10-gene risk score was significantly associated with recurrence risk of HR+/HER2- EBC after C-ET and effectively distinguished between high-risk and low-risk patients (training: HR: 6.37, P < 0.001; validation: HR: 4.51, P < 0.001). It maintained consistent stratification efficacy across different treatment regimens, clinical stages, and grades. Compared to existing multi-gene signatures (21-gene, 70-gene, EndoPredict, PAM50, GGI), HR+/HER2- EBC patients identified as high-risk by the 10-gene risk score exhibited a higher 10-year cumulative recurrence rate following C-ET. In multivariate Cox regression analysis, the 10-gene risk score remained an independent prognostic factor in both the training and validation sets. The comprehensive model, integrating the 10-gene score and clinicopathological features, showed high predictive accuracy (AUC: 0.734, 0.778, 0.792 for 3, 5, 7 years in training; 0.691, 0.715, 0.709 in validation).

CONCLUSION

The 10-gene risk score can serve as a tool to predict recurrence risk in HR+/HER2- EBC patients following C-ET, assisting clinicians in developing personalized treatment plans for high-risk patients and ultimately improving patient prognosis.